Issue 001 · 10 September 2026 · 5 min read · Operations & Delivery
This week’s question
Are we adopting AI — or actually redesigning work?
The technology is moving quickly. The more important question is whether the way we work is changing with it.
Why we're asking it
AI adoption in UK businesses is accelerating.
The Office for National Statistics says the proportion of businesses with 10 or more employees reporting use of at least one AI technology has almost tripled since late 2023.
In June 2026, 29% of businesses reported using at least one form of AI technology. Among larger businesses, the figure was 49%.
Text generation was the most commonly reported technology, followed by visual content creation.
There is clearly no shortage of interest.
And there is no shortage of expectation either.
In Deloitte's latest UK CFO survey, 93% of finance leaders expected investment in digital technology to increase over the next 12 months, while 73% were optimistic that AI would materially improve business performance.
So perhaps the question is no longer: will businesses adopt AI? They already are.
The more useful question might be: are businesses changing how work actually gets done — or simply adding AI to the way they already work?
What we're seeing
There is a significant difference between using AI and operating differently because of AI.
A team can use AI to write an email faster. Someone can summarise a meeting in seconds. A presentation can be created more quickly. Research that previously took an hour might take ten minutes.
All of that can be useful.
But it doesn't necessarily mean the organisation itself has become significantly more productive.
Government research published this year found that among UK businesses using AI, only around one in five said their AI tools were integrated into their existing business systems.
ONS analysis also found that, while AI adoption has grown strongly, the average number of AI technologies being used by adopting businesses has risen only modestly.
That suggests something important. Many organisations may still be in the tool adoption stage rather than the operating-model change stage.
And those are very different things.
Adding technology to a bad process doesn't make it a good process
This isn't really an AI problem. We've seen the same pattern with technology for years.
A business has an inefficient process. A new platform arrives. Instead of asking whether the process itself should exist or work differently, the existing process gets digitised.
The result? The business now has an inefficient process running through better technology.
AI creates exactly the same risk — only faster.
If five approvals aren't necessary, using AI to prepare the paperwork for those five approvals doesn't fix the problem.
If information is entered into three different systems, automatically moving it between them may save time, but perhaps the better question is why three systems are needed.
If a weekly report takes four hours to produce and nobody makes a decision from it, generating the report in four minutes isn't necessarily a productivity breakthrough.
You've saved three hours and 56 minutes. But eliminating a report nobody needs could save all four.
That distinction matters.
Before reaching for a technology solution, it is worth making sure you're solving the right problem.
Three things worth thinking about
1. Start with the work, not the AI
Instead of asking where we could use AI, try asking where work currently takes too long, requires unnecessary effort or depends too heavily on manual intervention.
Look at the work first.
- — Where do people repeatedly copy information?
- — Where are decisions waiting for someone?
- — Where do customers have to chase?
- — Where are employees searching for information that already exists somewhere else?
- — Where are highly capable people spending time on low-value administration?
2. Time saved isn't productivity unless something happens with it
Imagine AI saves an employee five hours a week. Excellent. What happens to those five hours?
If nobody knows, the organisation may technically have created capacity without turning it into anything valuable. The benefit comes when that capacity is deliberately redirected.
Five hours could become more customer conversations, better quality control, faster delivery, more commercial activity, better coaching, product development, problem solving, or simply the ability to handle growth without immediately adding headcount.
AI doesn't decide that. Leadership does.
So measuring adoption isn't enough. We also need to ask what we are doing with the capacity it creates.
3. The biggest opportunity may sit across a process, not within a role
Most early AI adoption naturally happens person by person. Someone discovers a tool. They find a useful application. They work a little faster. That's valuable.
But the bigger opportunity often appears when you stop thinking about individual tasks and look at an entire workflow.
Consider a customer enquiry. It might pass through Marketing, Sales, Operations, Finance, Delivery and Customer Support.
Improving one person's email-writing speed is useful. Understanding how AI, automation, better data and better process design could remove unnecessary handovers across that entire journey is potentially much more valuable.
That requires operations thinking, not simply technology adoption.
The productivity question
This matters particularly because productivity remains one of the UK's persistent business challenges.
The latest ONS estimate shows UK output per hour was 0.7% higher in the year to Q2 2026.
Growth is growth. But it puts the scale of the AI opportunity into perspective.
Businesses are investing heavily in technology partly because they expect it to improve productivity. To realise that promise, we may eventually need to move beyond giving existing roles better tools.
We may need to redesign how work flows, where decisions happen, what information people need, which activities genuinely add value, what gets automated, what stays human, and what no longer needs doing at all.
That's a much bigger conversation than choosing an AI platform.
Ask this in your business this week
Take one recurring process in your organisation. Not your entire company. Just one. Something your team does every week.
Then ask:
If we were designing this process from scratch today, knowing what technology can now do, would we design it this way?
If the answer is no, don't immediately buy another tool. Ask what you would change first.
You might discover that the biggest productivity opportunity has very little to do with AI itself.
The takeaway
The businesses that gain the most from AI may not be the ones that adopt the most tools. They may be the ones prepared to question the work around them.
- Where does effort go?
- What creates value?
- What slows people down?
- What should technology do?
- What should people do?
- And what shouldn't anyone be doing anymore?
AI gives businesses an extraordinary new set of capabilities. But technology alone doesn't redesign an organisation. People do.
So perhaps the question isn't how much AI are we using. It's: what are we now able to do differently?
Everything starts with a question.
Question. Solve. Deliver.
Sources
- Office for National Statistics — Artificial intelligence in UK businesses: 2023 to 2026 (20 July 2026)
- UK Business Data Survey 2026 — Department for Science, Innovation and Technology
- Office for National Statistics — Productivity flash estimate and overview, UK: April to June 2026 (18 August 2026)
- Deloitte UK CFO Survey Q2 2026 (20 July 2026)
What could your business do differently?
Technology can create opportunities. Real improvement comes from understanding where work, processes and decisions should change around it. OpsQ works alongside businesses to identify those opportunities and turn them into practical improvements.
Talk to OpsQ about the problemPrevious edition
This is the first edition.
Get The Q
One question. One useful perspective. Once a week.
Practical thinking on operations, business, growth, leadership and work — straight to your inbox.